The 186-4 RSA Validation System (RSA2VS)

Page created by Mike Brewer
 
CONTINUE READING
The 186-4 RSA Validation System (RSA2VS)

                     Updated: July 8, 2014
                    Previous: April 16, 2014
                    Original: March 03, 2011

                        Sharon S. Keller

          National Institute of Standards and Technology

               Information Technology Laboratory

                   Computer Security Division

          Cryptographic Algorithm Validation Program
Update Log
                                               7/8/14

   Fix typo in Section 6.2.2 – instead of sentences saying “if i is not a prime”, changed to “if p is not
    a prime” and “if q is not a prime”.

                                              4/16/14

   Update Section 6.2 to indicate the IUT supplies all the values for Key Generation.

                                              3/18/14

   Update Key Gen section to reflect change in way testing performed. Now the IUT supplies all
    the values instead of the CAVS supplying some initial values first.

   Moved the 186-2 Legacy Signature Verifications tests to this document from the 186-2 RSAVS
    document

                                               9/5/13

   Updated References to FIPS 186-4 throughout document.

                                              10/25/12

   Section 6.2.1 Clarified what the IUT is responsible for providing for four of the methods of Key
    Generation that are specified in FIPS186-4 Appendices B.3.2, B.3.4, B.3.5, and B.3.6.

                                              3/22/12

   Section 6.2.2 Validation Test for Key Generation Methods Specified in Appendices B.3.3

        o   Distinguished the testing requirements for IUTs supporting only fixed e value(s) vs. IUTs
            supporting random e values.

   Section 6.2.2.1 The Known Answer Test for B.3.3 Probably Primes

        o   Clarified that this test is only for use with IUTs supporting random e values.

   Section 6.2.2.2 The KeyGen_RandomProbablyPrime3_3 Test

        o   Clarified that this test is required for both IUTs supporting random e values and IUTs
            supporting fixed e values.

                                              6/29/11

                                                  ii
   Section 2

       o   Removed second paragraph on testing of prerequisites.

   Section 4

       o   Removed DRBG, DRBGVS, SHA and SHAVS from list of abbreviations.

   Section 6

       o   Removed second paragraph on testing of prerequisites.

                                            iii
TABLE OF CONTENTS

1      INTRODUCTION ..............................................................................................................................................1
2      SCOPE .................................................................................................................................................................1
3      CONFORMANCE ..............................................................................................................................................2
4      DEFINITIONS AND ABBREVIATIONS ........................................................................................................2
5      DESIGN PHILOSOPHY OF THE 186-4RSA VALIDATION SYSTEM .....................................................2
6      186-4RSAVS TESTS ...........................................................................................................................................3
    6.1     CONFIGURATION INFORMATION....................................................................................................................3
    6.2     186-4 RSA KEY GENERATION TEST .............................................................................................................4
       6.2.1 Validation Test for Key Generation Methods Specified in Appendices B.3.2, B.3.4, B.3.5, and B.3.6 ....4
       6.2.2 Validation Test for Key Generation Methods Specified in Appendices B.3.3 ..........................................7
    6.3     186-4 RSA SIGNATURE GENERATION TEST ................................................................................................ 11
    6.4     186-4 RSA SIGNATURE VERIFICATION TEST .............................................................................................. 13
    6.5     186-2 LEGACY RSA SIGNATURE VERIFICATION TEST ................................................................................ 15
APPENDIX A. REFERENCES ................................................................................................................................ 17

                                                                                     iv
1       Introduction
The 186-4 RSA Validation System (RSA2VS) specifies the procedures involved in validating
implementations of public key cryptography based on the RSA algorithm. This document deals with
three variations of the RSA algorithm which are referenced in FIPS186-4, Digital Signature
Standard (DSS) [1]. They are:

    • RSA algorithm specified in American National Standard (ANS) X9.31 [2], and

    • Two signature schemes with appendix specified in Public Key Cryptography Standards(PKCS)
       #1 v2.1: RSA Cryptography Standard-2002 [3]. These two signature schemes with appendix
       are

            o RSASSA-PSS, and

            o RSASSA-PKCS1-v1_5.

The 186-4RSAVS is designed to perform automated testing on Implementations Under Test (IUTs).
It provides the basic design and configuration of the 186-4RSAVS. Included are the specifications
for testing the individual RSA components of the IUT. These components include:

       Key Generation

       Signature Generation

       Signature Verification

       FIPS 186-2 Legacy testing for Signature Verification

This document defines the purpose, the design philosophy, and the high-level description of the
validation process for RSA. The requirements and administrative procedures to be followed by those
seeking formal validation of an implementation of 186-4RSA are presented. The requirements
described include the specification of the data communicated between the IUT and the 186-4RSAVS,
the details of the tests that the IUT must pass for formal validation, and general instruction for
interfacing with the 186-4RSAVS.

2       Scope
This document specifies the tests required to validate IUTs for conformance to the 186-4RSA as
specified in [1]. When applied to IUTs that implement any of the three different algorithm variations
of the RSA, the 186-4RSAVS provides testing to determine the correctness of the algorithm
components contained in the implementation. The 186-4RSAVS is composed of three different tests,
one to validate each of the various algorithm components and the 186-2 Legacy tests for Signature
Verification. In addition to determining conformance to the cryptographic specifications, the 186-

                                                  1
4RSAVS is structured to detect implementation flaws including pointer problems, insufficient
allocation of space, improper error handling, and incorrect behavior of the RSA implementation.

3      Conformance
The successful completion of the tests contained within the 186-4RSAVS is required to be validated
as conforming to the RSA. Testing for the cryptographic module in which the RSA is implemented is
defined in FIPS PUB 140-2, Security Requirements for Cryptographic Modules.[4]

4      Definitions and Abbreviations
DEFINITION                                        MEANING

CST laboratory                                    Cryptographic and Security Testing laboratory
                                                  that operates the 186-4RSAVS

IUT                                               Implementation Under Test

PKCS                                              Public Key Cryptography Standards

5      Design Philosophy of the 186-4RSA Validation System
The 186-4RSAVS is designed to test conformance to RSA rather than provide a measure of a
product’s security. The validation tests are designed to assist in the detection of accidental
implementation errors, and are not designed to detect intentional attempts to misrepresent
conformance. Thus, validation should not be interpreted as an evaluation or endorsement of overall
product security.

The 186-4RSAVS has the following design philosophy:

       1. The 186-4RSAVS is designed to allow the testing of an IUT at locations remote to the 186-
               4RSAVS. The 186-4RSAVS and the IUT communicate data via REQUEST and
               RESPONSE files.

       2. The testing performed within the 186-4RSAVS utilizes statistical sampling (i.e., only a
               small number of the possible cases are tested); hence, the successful validation of a
               device does not imply 100% conformance with the standard.

                                                  2
6        186-4RSAVS Tests
The 186-4RSAVS for RSA provides conformance testing for three of the components of the
algorithm, as well as testing for apparent implementation errors. The components tested are key
generation as specified in FIPS186-4, signature generation and signature verification. There are 5
different methods used for key generation which are discussed in FIPS 186-4, Appendix B.3.

The 186-4RSAVS also contains 186-2 Legacy testing for the Signature Verification

6.1      Configuration Information

To initiate the validation process of the 186-4RSAVS, a vendor submits an application to an
accredited laboratory requesting the validation of its implementation of 186-4 RSA. The vendor’s
implementation is referred to as the Implementation Under Test (IUT). The request for validation
includes background information describing the IUT along with information needed by the 186-
4RSAVS to perform the specific tests. More specifically, the request for validation includes:

      1. Vendor Name;

      2. Product Name;

      3. Product Version;

      4. Implementation in software, firmware, or hardware;

      5. Processor and Operating System with which the IUT was tested if the IUT is implemented in
          software or firmware;

      6. Brief description of the IUT or the product/product family in which the IUT is implemented by
          the vendor (2-3 sentences); and

      7. For RSA from X9.31 and RSASSA_PKCS1.5, the modulus size-SHA size combination(s)
          supported by the IUT.

      8. For RSASSA-PSS implementations, the modulus size-SHA size-SALT length and, if
          applicable, the SALT value combination(s) supported by the IUT.

      9. For IUT’s implementing Key Generation,

             a. The prime methods supported by the IUT. Each prime method requires some
                combination of the SHA sizes supported, the modulus sizes supported and/or the
                table used to determine the number of iterations of the Miller-Rabin routine.

                                                   3
6.2      186-4 RSA Key Generation Test

An implementation of the FIPS 186-4 RSA may generate the key components used in the 186-4 RSA
algorithm’s signature generation and verification processes. This option tests the ability of an IUT to
produce correct values for these components.

FIPS186-4 Key Pair generation specifies 5 ways (or methods) to generate the primes p and q. These
include:

      1. Random Primes:
              Provable primes (Appendix B.3.2)
              Probable primes (Appendix B.3.3)
      2. Primes with Conditions:
              Primes p1, p2, q1,q2, p and q shall all be provable primes (Appendix B.3.4)
              Primes p1, p2, q1, and q2 shall be provable primes and p and q shall be probable
                primes (Appendix B.3.5)
              Primes p1, p2, q1,q2, p and q shall all be probable primes (Appendix B.3.6)

The Random Provable primes (Appendix B.3.2), and all the Primes with conditions (Appendix B.3.4,
B.3.5, and B.3.6) can be tested in the same manner because each of these begin with a starting
random number and calculate the p and q values from this value. The 186-4 RSAVS instructs the
IUT to generate intermediate values and the p, q, n, and d values. The RSAVS then validates the
correctness of the values generated by the IUT.

The Random Probable primes (Appendix B.3.3) must be tested in a different way because this test
generates different random numbers, not related to each other, until the number satisfies the
“probably prime” requirements. The 186-4 RSAVS validation requires two tests for Random
Probable primes. These include the Known Answer Test and the
KeyGen_RandomProbablyPrime3_3 test.

6.2.1 Validation Test for Key Generation Methods Specified in Appendices B.3.2,
B.3.4, B.3.5, and B.3.6

To test the key generation method specified in Appendices B.3.2, B.3.4, B.3.5, and B.3.6, the 186-4
RSAVS requests the IUT to supply 25 sets of values for each prime method/modulus/hash(if
applicable)/Miller-Rabin Table used(if applicable) combination supported by the IUT. The 186-4
RSAVS verifies the received results by calculating the values based on the values used by the IUT.
This will assure that the IUT can generate the primes using the key generation method specified.

The 186-4 RSAVS:
                                                   4
A. Creates a REQUEST file (Filename: KeyGen_186-3.req) containing:
                    1. A header consisting of:
                          a. The CAVS version,
                          b. The algorithm being tested and the product name,
                          c. For each prime method selected (Appendices B.3.2, B.3.4, B.3.5, and
                              B.3.6):
                                 i.   The name of the prime method selected,
                                ii.   The modulus or modulus/hash combinations supported,
                               iii.   If applicable, the Miller-Rabin information; and
                          d. The date generated.
                    2. For each prime method – modulus size – hash algorithm (if applicable) –
                        Miller-Rabin table combination (if applicable) tested,
                           a. a section header indicating the value for each the prime method, the
                                 modulus size, the hash algorithm (if applicable), and the Miller-
                                 Rabin table (if applicable).
                           b. N = 25 indicating the number of entries the IUT is to send in the
                                 response file.
                       Note: The CST laboratory sends the REQUEST file to the IUT.

      B. Creates a FAX file (Filename: KeyGen_186-3.fax) containing:
              1. The information from the REQUEST file.
The IUT:

   A. Produces N sets of values per prime method, the modulus size, the hash algorithm (if
       applicable), and the Miller-Rabin table (if applicable) combination supported by the IUT.
   B. Creates a RESPONSE file (Filename: KeyGen_186-3.rsp) containing:
           a. 25 sets of values per prime method, the modulus size, the hash algorithm (if
              applicable), and the Miller-Rabin table (if applicable) combination supported by the
              IUT. The values in each set differs for each prime method:
                  i. Prime method: Provable Random primes (Appendix B.3.2)
                          1. Public key, e,
                          2. Seed
                          3. The private prime factor p
                          4. The private prime factor q
                          5. The value of the modulus n
                          6. The value of the private signature exponent d.

                  ii. Prime method: Primes with Conditions where p1, p2, q1,q2, p and q shall all
                      be provable primes (Appendix B.3.4)

                                                 5
1.   Public key,e,
        2.   Seed,
        3.   Bitlen1 for length of p1
        4.   Bitlen2 for length of p2
        5.   Bitlen3 for length of p3
        6.   Bitlen4 for length of p4
        7.   The private prime factor p
        8. The private prime factor q
        9. The value of the modulus n
        10. The value of the private signature exponent d.
iii. Prime method: Primes with Conditions where p1, p2, q1, and q2 shall be
     provable primes and p and q shall be probable primes (Appendix B.3.5)
         1. Public key, e,
         2. Seed,
         3. Bitlen1 for length of p1
         4. The prime factor p1
         5. Prime seed for p2
         6. Bitlen2 for length of p2
         7. The prime factor p2
         8. Prime seed for q1
         9. Xp – the random number used in Step 3 of Appendix C.9 Compute a
             Probable Prime Factor Based on Auxiliary Primes to generate prime
             p
         10. The private prime factor p
         11. Bitlen3 for length of p3
         12. The prime factor q1
         13. Prime seed for q2
         14. Bitlen4 for length of p4
         15. The prime factor q2
         16. Xq– the random number used in Step 3 of Appendix C.9 Compute a
             Probable Prime Factor Based on Auxiliary Primes to generate prime
             q
         17. The private prime factor q
         18. The value of the modulus n
         19. The value of the private signature exponent d .

iv. Prime method: Primes with Conditions where p1, p2, q1,q2, p and q shall all
    be probable primes (Appendix B.3.6)
        1. Public key, e,
        2. Bitlen1 for length of Xp1

                               6
3.    Xp1 – Random odd integer for Step 4.1 to calculate p1
                            4.    The prime factor p1
                            5.    Bitlen2 for length of Xp2
                            6.    Xp2 – Random odd integer for Step 4.1 to calculate p2
                            7.    The prime factor p2
                            8.    Xp – the random number used in Step 3 of Appendix C.9 to generate
                                  prime p
                            9.    The private prime factor p
                            10.   Bitlen3 for length of Xq1
                            11.   Xq1– Random odd integer for Step 4.1 to calculate q1
                            12.   The prime factor q1
                            13.   Bitlen4 for length of Xq2
                            14.   Xq2 – Random odd integer for Step 4.1 to calculate q2
                            15.   The prime factor q2
                            16.   Xq – the random number used in Step 3 of Appendix C.9 to generate
                                  prime q
                            17.   The private prime factor q
                            18.   The value of the modulus n
                            19.   The value of the private signature exponent d.

Note: The IUT sends the RESPONSE file to the CST laboratory for processing by the 186-4 RSAVS.
The 186-4 RSAVS:

       A.       Using the supplied values, the RSAVS computes all values and compares them to the
                   values in the response file.
       B.       If all the values pass, the test is successful. If any of the values do not pass, the test
                   will fail.

6.2.2 Validation Test for Key Generation Methods Specified in Appendices B.3.3
To test the key generation method specified in Appendices B.3.3, the IUT will perform either one or two
tests depending on whether or not the IUT supports random e values. If the IUT supports random e
values, the IUT will be instructed to perform both the Known Answer Test (Section 6.2.2.1) and the
KeyGen_RandomProbablyPrime3_3 test (Section 6.2.2.2).

6.2.2.1         The Known Answer Test for B.3.3 Probably Primes
The Known Answer Test will supply the IUT with multiple iterations of public key, e, and random
numbers for p and, if applicable, q. These values are supplied for each combination of modulus size and
Miller-Rabin Table used. These values will either be successful in obtaining primes p and q or will fail.
The values that are successful get to Step 5.6.2 Q is PROBABLY PRIME – Result = SUCCESS. Note to
get to Step 5.6.2, the p value supplied is PROBABLY PRIME and has passed Step 4.5.2. The values that
fail will fail at one of several locations in the B.3.3 function. Possible reasons for failure include:
                                                    7
Step 4.4 If ((p < (√2 )(2(nlen / 2) – 1)), then go to Step 4.2 and generate another random number.

Step 4.7 If (i < (5*(nlen/2)) and p is not prime go to Step 4.2 and generate another random number.

Step 5.5 If ((q < (√2 )(2(nlen / 2) – 1)), then go to Step 5.2 and generate another random number.

Step 5.8 If (i < 5*(nlen/2)) and q is not prime then go to Step 5.2 and generate another random number.

Note that the KAT generates random e values and other inputs to assure that all possible errors listed
above are tested by the IUT. If an IUT supports only fixed e values, this IUT may not be able to process
this file. Therefore this test is not required for IUTs only supporting fixed e value(s).

The 186-4 RSAVS:

Creates a REQUEST file (Filename:KeyGen3_3_KAT.req) containing:

A header consisting of:

       The CAVS version

       The algorithm being tested and the product name,

       The modulus supported,

       The Miller-Rabin information

       The date generated.

       For each modulus size/Miller-Rabin table combination tested,

       A section header indicating the value for each modulus size, and the Miller-Rabin table.

       Groups of data including

             o   Public key, e

             o   Random number for p

             o   Random number for q (if applicable)

Note: The CST laboratory sends the REQUEST file to the IUT.

B. Creates a FAX file (Filename: KeyGen3_3_KAT.fax) containing:

        1. The information from the REQUEST file and

        2. Result = P or F failure code. The failure code is in the form of the instruction step number
          from FIPS186-4 Appendix B.3.3 (see above).

The IUT:
                                                        8
A. Uses the mod size, the public key exponent e and the random number for p, and, if applicable, q to
determine if the random number is PROBABLY PRIME.

B. Creates a RESPONSE file (Filename: KeyGen3_3_KAT.rsp) containing:

        1. The information from the REQUEST file and

        2. Result = P or F.

Note: The IUT sends the RESPONSE file to the CST laboratory for processing by the 186-4 RSAVS.

The 186-4 RSAVS:

A. Compares the contents of the RESPONSE file with the contents of the FAX file.

B. Records PASS for this test if the results for all public key e, private prime factor p, private prime factor
q, and Result match; otherwise, records FAIL.

6.2.2.2          The KeyGen_RandomProbablyPrime3_3 Test

The KeyGen_RandomProbablyPrime3_3 test for Appendix B.3.3 consists of the 186-4 RSAVS
asking the IUT to supply 10 prime p, q pairs for each modulus size/Miller-Rabin table combination
supported by the IUT. This will assure that the IUT can generate correct probable primes using
Appendix B.3.3. This test is required for both IUTs supporting fixed e value(s) AND for IUTs supporting
random e values.

The 186-4 RSAVS:

        A. Creates a REQUEST file (Filename:KeyGen_RandomProbablyPrime3_3.req)
           containing:
              1. A header consisting of:
                        i. The CAVS version
                       ii. The algorithm being tested and the product name
                      iii. The modulus supported
                      iv. The Miller-Rabin information
                       v. The date generated.
              2. For each modulus size/Miller-Rabin table combination tested,
                        i. A section header indicating the value for each modulus size, and the
                           Miller-Rabin table.
                       ii. N = 10 indicating the number of entries the IUT is to send in the response
                           file.

                Note: The CST laboratory sends the REQUEST file to the IUT.
        B. Creates a FAX file (Filename: KeyGen_RandomProbablyPrime3_3.fax) containing:
                                                       9
1. The information from the REQUEST file.
The IUT:

   A. Produces N sets of values per modulus size/Miller-Rabin table combination supported by the
IUT.
   B. Creates a RESPONSE file (Filename: KeyGen RandomProbablyPrime3_3.rsp) containing:
       a.     10 sets of values per modulus size/Miller-Rabin table combination supported by the
              IUT. The values in each set includes:
                        i.  e, the public key exponent,
                       ii.  p, the probable prime,
                     iii.   q, the probable prime,
                      iv.   n, and
                       v.   d, the private key exponent.

Note: The IUT sends the RESPONSE file to the CST laboratory for processing by the 186-4 RSAVS.
The 186-4 RSAVS:

       A.     Using the e, p, and q, the RSAVS runs Appendix B.3.3 using p in Section 4.2 and q
                in Section 5.2 to assure that each number is a probable prime.
       B.     Compute n and d and compare them to the values in the response file.
       C.     If all the values pass, the test is successful. If any of the values do not pass, the test
              will fail.

                                                   10
6.3    186-4 RSA Signature Generation Test
An implementation of FIPS 186-4 RSA may generate the digital signature. There are 3 different RSA
algorithms included in FIPS 186-4. They include the RSA from X9.31, the PKCS#1 1.5, and the
PKCS#1 PSS algorithms. FIPS 186-4 introduces additional constraints for each of these RSA
algorithms.

The FIPS 186-4 RSA Signature Generation Test tests the ability of an IUT to produce correct
signatures. To test signature generation, the 186-4RSAVS supplies ten messages to the IUT for each
modulus size/SHA algorithm combination. The IUT generates the corresponding signatures and
returns them to the 186-4RSAVS. The 186-4RSAVS validates the signatures by using the associated
public key to verify the signature.

The 186-4RSAVS:

       A. Creates a REQUEST file and a FAX file for each RSA algorithm being tested, (Request
              Filenames: SigGen931_186-3.req, SigGen15_186-3.req, SigGenPSS_186-3.req)
              (FAX Filenames: SigGen931_186-3.fax, SigGen15_186-3. fax, SigGenPSS_186-3.
              fax) containing:
              1. A header consisting of:
                      a. The CAVS version,
                      b. The algorithm being tested and the product name,
                      c. The modulus size/SHA size (with SALT len if applicable) combinations,
               and
                      d. The date generated.
              2. For each modulus size tested, a section header indicating the modulus length
              3. Ten groups of data for each modulus size/SHA size combination supported
              consisting of
                      a. The SHA algorithm supported,
                      b. A message to be signed.

              Note: The CST laboratory sends the REQUEST file to the IUT.
The IUT:

       A. Generates the signatures for the messages supplied in the REQUEST file.

                                                11
B. Creates a RESPONSE file for each RSA algorithm being tested (Request Filenames:
      SigGen931_186-3. rsp, SigGen15_186-3. rsp, SigGenPSS_186-3. rsp) containing:
             1. The header described above,
             2. For each mod size tested,
                  a. a section header indicating the mod length,
                  b. a modulus, n,
                  c. a public key, e, corresponding to the private key, d, used to generate the
                     signatures,
             3. Ten groups of data for each SHA algorithm supported consisting of
                  a. The SHA algorithm supported,
                  b. A message, Msg,
                  c. Signature value, s.

             Note: The IUT sends the RESPONSE file to the CST laboratory for processing by the
             186-4RSAVS.
The 186-4RSAVS:

      A. Uses the respective public keys to verify the signatures in the RESPONSE file.
      B. Records PASS for this test if all conditions are met; otherwise, records FAIL.

                                                 12
6.4    186-4 RSA Signature Verification Test
An implementation of FIPS 186-4 RSA may verify the digital signature. There are 3 different RSA
algorithms included in FIPS 186-4. They include the RSA from X9.31, the PKCS#1 1.5, and the
PKCS#1 PSS algorithms. FIPS 186-4 introduces additional constraints for each of these RSA
algorithms.

The FIPS 186-4 RSA Signature Verification Test tests the ability of the IUT to recognize valid and
invalid signatures. For each modulus size/SHA algorithm selected, the 186-4RSAVS generates a
modulus and three associated key pairs, (d, e). Each private key d is used to sign six pseudorandom
messages each of 1024 bits. Some of the public keys, e, messages, IR format, or signatures are
altered so that signature verification should fail. The modulus, SHA algorithm, public key e values,
messages, and signatures are forwarded to the IUT. The IUT then attempts to verify the signatures
and returns the results to the 186-4RSAVS, which compares the received results with its own stored
results.

The 186-4RSAVS:

       A. Generates 3 groups of data for each supported modulus size. Each group consists of a
             modulus and 6 sets of data for each supported SHA algorithm. Each set of data
             contains

               1. The SHA algorithm supported,

               2. A public/private key pair that is consistent with the modulus,

               3. A pseudorandom message and

               4. A signature for the message using the private key.

       B. Alters the public key e, the message, the IR format, or the signature for five out of six of
               the public key/message/signature sets such that the message verification fails.

       C. Creates a REQUEST file (Filename: SigVer931_186-3.req, SigVer15_186-3.req,
              SigVerPSS_186-3.req) containing:
               1.      A header consisting of:
                       a. The CAVS version,
                       b. The algorithm being tested and the product name,
                       c. The modulus size/SHA size (with SALT len if applicable) combinations,
               and

                                                  13
d. The date generated.
             2.      For each modulus size supported:
                     a. A section header indicating the modulus length
                     b. The modulus n for the supported modulus size,
                     c. 3 groups of data consisting of 6 sets of data for each supported SHA
                         algorithm. Each set of data contains:
                               i. The SHA algorithm supported,
                               ii. A public key e corresponding to the private key used to sign the
                                  messages,
                               iii. The pseudorandom message and
                               iv. The signature.

             Note: The CST laboratory sends the REQUEST file to the IUT.
      D. Creates a FAX file (Filename: SigVer931_186-3.fax, SigVer15_186-3. fax,
      SigVerPSS_186-3.fax) containing:
             1. The information from the REQUEST file,
             2. The private key, d, associated with the public key, e, and
             3. The result indicating whether the signature verification process should pass or
                fail, for each public key/message/signature set, and, if it should fail, why.
The IUT:

      A. Attempts to verify the signatures for the messages supplied in the REQUEST file using the
             corresponding modulus n, SHA algorithm and the public key e.
      B. Creates a RESPONSE file (Filename: SigVer931_186-3.rsp, SigVer15_186-3.rsp,
      SigVerPSS_186-3.rsp) containing:
             1. The information from the REQUEST file and
             2. An indication of whether the signature verification passed or failed for each public
                     key/message/signature set.
             Note: The IUT sends the RESPONSE file to the CST laboratory for processing by the
             RSAVS.
The 186-4RSAVS:

      A. Compares the contents of the RESPONSE file with the contents of the FAX file.
      B. Records PASS for this test if the results for all public key/message/signature sets match;
      otherwise, records FAIL.
                                                14
6.5    186-2 Legacy RSA Signature Verification Test
This option tests the ability of the IUT to recognize valid and invalid signatures. For each modulus
size/SHA algorithm selected, the RSAVS generates a modulus and three associated key pairs, (d, e).
Each private key d is used to sign four pseudorandom messages each of 1024 bits. Some of the
public keys, e, messages or signatures are altered so that signature verification should fail. The
modulus, SHA algorithm, public key e values, messages, and signatures are forwarded to the IUT.
The IUT then attempts to verify the signatures and returns the results to the RSAVS, which compares
the received results with its own stored results.

The RSAVS:

       A.      Generates 3 groups of data for each supported modulus size. Each group consists of a
               modulus and 4 sets of data for each supported SHA algorithm. Each set of data
               contains

               1.     The SHA algorithm supported,

               2.     A public/private key pair that is consistent with the modulus,

               3.     A pseudorandom message and

               4.     A signature for the message using the private key.

               For the efficiency of the tool, a modulus will sometimes be used for more than one
               group of data when it is consistent with more than one public key. Therefore when
               loading the modulus for each group of data, the modulus specified for this group may
               be the same as that specified for the previous group.

       B.      Alters the public key e, the message or the signature for three fourths of the public
               key/message/signature sets such that the message verification fails.

       C.      Creates a REQUEST file (Filename: SigVer931_186-2.req, SigVer15_186-2.req,
               SigVerPSS_186-2.req) containing:
       1.      The Product Name;
               2.     For each modulus size supported:
                       a. The modulus n for the supported modulus size,
                       b. 3 groups of data consisting of 4 sets of data for each supported SHA
                          algorithm. Each set of data contains:
                                i. The SHA algorithm supported,
                               ii.   The information from step B, including:
                                        1. A public key e corresponding to the private key used to
                                           sign the messages,
                                        2. The pseudorandom message and
                                                  15
3. The signature.
              Note: The CMT laboratory sends the REQUEST file to the IUT.
       D.    Creates a FAX file (Filename: SigVer931_186-2.fax, SigVer15_186-2. fax,
SigVerPSS_186-2. fax) containing:
              1.      The information from the REQUEST file and
              2.     An indication of whether the signature verification process should pass or fail,
                     for each public key/message/signature set.
The IUT:

       A.     Attempts to verify the signatures for the messages supplied in the REQUEST file
              using the corresponding modulus n, SHA algorithm and the public key e.
       B.    Creates a RESPONSE file (Filename: SigVer931_186-2.rsp, SigVer15_186-2. rsp,
SigVerPSS_186-2. rsp) containing:
       1.     The information from the REQUEST file and
              2.     An indication of whether the signature verification passed or failed for each
                     public key/message/signature set.
              Note: The IUT sends the RESPONSE file to the CMT laboratory for processing by the
              RSAVS.
The RSAVS:

       A.     Compares the contents of the RESPONSE file with the contents of the FAX file.
       B.     Records PASS for this test if the results for all public key/message/signature sets
              match; otherwise, records FAIL.

                                                 16
Appendix A. References
[1] FIPS186-4, Digital Signature Standard (DSS), FIPS Publication 186-4, National Institute of
       Standards and Technology, July, 2013.

[2] American National Standard (ANS) X9.31 Digital Signatures Using Reversible Public Key
      Cryptography for the Financial Services Industry (rDSA), September, 1998.

[3] PKCS#1 v2.1: RSA Cryptography Standard, RSA Laboratories, June 14, 2002.

[4] Security Requirements for Cryptographic Modules, FIPS Publication 140-2, National Institute of
Standards and Technology, May 2001.

                                                 17
You can also read